Abstract We propose to develop an integrated system for positron emission tomography (PET) and electron paramagnetic resonance (EPR) simultaneous imaging of rodents for use in studying cancer and cancer treatment. Cancer is a complex and heterogeneous functional disease. Therefore, effective treatment of cancer shall account for the specific cellular-type populations and their states, as well as the tumor microenvironment. In vivo imaging has an important role in providing such information. PET, already widely used in the clinic, can measure many metabolic and physiological states of tumor, including glucose utilization with 18F-fluorodeoxyglucose (FDG) and cell proliferation with 18F-fluorothymidine (FLT). PET tracers for imaging tumor specific cell-membrane or intracellular molecules or genes, and for imaging tumor microenvironment including angiogenesis and extracellular matrix, are also available or under active development. The partial oxygen pressure, pH and interstitial inorganic phosphate in the space surrounding the cancer cells also shall be considered as they will affect metabolic and physiological processes, and hence the treatment outcome. Particularly, knowing the tissue oxygen level is of importance for cancer treatment. It has been well known that hypoxic tumor cells are resistant to various therapeutic agents. Therefore, many believe that dose painting in which the radiation dose to the tumor is adjusted according to the local tissue oxygen level can improve cancer cure. Our recent data in mice shows that performing dose painting based on EPR oxygen maps can reduce the radiation dose to well-oxygenated areas of the tumor by 30%, resulting in not only improved treatment outcome but also reduced post treatment risks and complications. To date, EPR imaging is the only proven method for absolute in vivo measurement of the tissue oxygen level. While PET hypoxia imaging with 18F-Misonidazole (FMISO) has been investigated, FMISO uptake depends on many factors other than tissue oxygen concentration in nontrivial ways. This can explain why similar positive results with dose painting have not been observed in the clinic when performed based on PET-FMISO images. The proposed integrated system will be a powerful research tool for studying many heterogenous functional abnormalities in tumor and for developing and improving cancer treatment. In this project, we also will employ the developed system to conduct a small pilot study for purpose of validating the system for real animal imaging and demonstrating its potential usefulness for studying the correlation between PET-FMISO images and EPR oxygen maps. Such correlation, when identified, can lead to the development of methods for correcting the PET-FMISO images such that they can be successfully employed for dose painting to yield improved cancer cure. If successful, the clinical impact will be significant and immediate as FMISO-PET imaging can be readily employed in the clinic.